No, they utilized a more (currently) traditional multiple-render-pass process to have individual control over key lights, ambient lights, fill, reflection, highlights, occlusion, etc. This gives the compositors more control over the final look. The other side of the issue is that it also makes knowing what it would "really" look like a bit more complex to get to, since you have almost limitless control to push things easily past what would be physically plausible, given that the environments - digital or not - give off and bounce light of varying quality. That's why you need talented compositors, who have an eye for that sort of thing. The next generation of rendering software basically gives you a real-world result, which of course you can alter, but at least you have a baseline that you can trust, and the brain instantly recognizes all the myriad subtle physical properties of light as being "correct". It's sort of complicated, but the bottom line is that it works.